Ship dynamics model identification method, system, device and medium

Abstract

The present invention discloses a ship dynamics model identification method, system, device and medium, and relates to the field of intelligent ship technology. The present invention calculates the cumulative probability density matrix of different types of ship data, thereby converting the distance similarity in the Euclidean space between the various types of ship data into the empirical similarity in the probability space, thereby emphasizing the similarity between the sparse values ​​of the ship data, and then combines the radial basis kernel algorithm to calculate the kernel function of the data pairs of multiple sample data, so that the high-dimensional fitting surface will not have too many abrupt values, making the identification of the future ship dynamics model more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent ships, and in particular to a ship dynamics model identification method, system, device and medium. Background Art

[0002] Ship maneuverability prediction helps to achieve reasonable control of the ship and reduce ship energy consumption. In recent years, ship motion control technology has been widely used in unmanned commercial ships and is a key technology for realizing autonomous ship transportation. In order to implement model-based ship motion control algorithms such as model predictive control (MPC), accurately control ship motion, and verify the reliability and robustness of the control algorithm, a high-precision ship dynamics model is required. Therefore, in the ship design stage, in order to ensure the excellent maneuverability of the ship and lay the foundation for the subsequent controller design, it is necessary to perform three-degree-of-freedom maneuvering prediction based on an accurate three-degree-of-freedom ship dynamics model. In related technologies, the motion prediction model is directly identified based on the relationship between the rudder angle and the heading, which has low accuracy. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a ship dynamics model identification method, system, device and medium, which can improve the accuracy of ship motion model identification.

[0004] On the one hand, an embodiment of the present invention provides a method for identifying a ship dynamics model, comprising the following steps:

[0005] Acquire a plurality of sample data at different time dimensions, wherein the sample data includes a plurality of ship data, and the plurality of ship data are divided into ship state quantity data and ship control quantity data;

[0006] Determine a cumulative probability density matrix of different types of ship data based on the plurality of sample data;

[0007] Based on the radial basis kernel algorithm, a kernel function is determined according to the cumulative probability density matrix of the plurality of ship data;

[0008] A least squares support vector machine is trained according to the kernel function and label data in the sample data to obtain a ship dynamics model, wherein the ship dynamics model is used to predict the ship state quantity data at the next moment.

[0009] According to some embodiments of the present invention, the ship data includes surge speed, sway speed, bow angular velocity, rudder angle and propeller speed, wherein the surge speed, the sway speed and the bow angular velocity are ship state quantity data, and the rudder angle and the propeller speed are ship control quantity data.

[0010] According to some embodiments of the present invention, determining the cumulative probability density matrix of different types of ship data according to the plurality of sample data comprises the following steps:

[0011] The same type of ship data from the sample data at different times is extracted to obtain the ship data set;

[0012] Determine the sequence number of each ship data in the ship data set according to the size order of each ship data in the ship data set;

[0013] The cumulative probability density of the ship data is determined according to the serial number of each ship data in the ship data set, and the cumulative probability density matrix of the ship data set is obtained.

[0014] According to some embodiments of the present invention, the kernel function of the data pair in the kernel function matrix is ​​expressed as:

[0015]

[0016] Among them, k(x,x') represents the kernel function of the data pair x and data x', P_CDF_x n -P_CDF_x' n It represents the degree of dissimilarity between corresponding data pairs in the cumulative probability density matrix of the nth type of ship data, N represents the number of ship data types, and a is the preset kernel width parameter.

[0017] According to some embodiments of the present invention, the least squares support vector machine is expressed as:

[0018]

[0019] Among them, k(x,x') represents the kernel function, m represents the number of sample data in the sample data, α i and b are parameters to be trained.

[0020] According to some embodiments of the present invention, parameter α i and b are solved by the following formula:

[0021]

[0022] Among them, C is the preset penalty factor, y i is the label data in the i-th sample data, and the label data is the ship state data at the next moment.

[0023] According to some embodiments of the present invention, the step of training a least squares support vector machine according to the kernel function and label data in the sample data to obtain a ship dynamics model comprises the following steps:

[0024] Determine that the tag data is the surge speed;

[0025] Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the surge velocities in the plurality of sample data to obtain a surge velocity prediction model;

[0026] Determining the tag data as the sway speed;

[0027] Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the sway velocities in the plurality of sample data to obtain a sway velocity prediction model;

[0028] Determine that the label data is the yaw angular velocity;

[0029] Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the sway velocities in the plurality of sample data to obtain a yaw angular velocity prediction model;

[0030] A three-degree-of-freedom ship dynamics model is determined according to the surge velocity prediction model, the sway velocity prediction model and the bow angular velocity prediction model.

[0031] On the other hand, an embodiment of the present invention further provides a ship dynamics model identification system, comprising:

[0032] The first module is used to obtain a plurality of sample data in different time dimensions, wherein the sample data includes a plurality of ship data, and the plurality of ship data are divided into ship state quantity data and ship control quantity data;

[0033] The second module is used to determine the cumulative probability density matrix of different types of ship data according to the plurality of sample data;

[0034] A third module is used to determine a kernel function according to the cumulative probability density matrix of the plurality of ship data based on a radial basis kernel algorithm;

[0035] The fourth module is used to train a least squares support vector machine according to the kernel function and the label data in the sample data to obtain a ship dynamics model, wherein the ship dynamics model is used to predict the ship state quantity data at the next moment.

[0036] On the other hand, an embodiment of the present invention further provides a ship dynamics model identification device, comprising:

[0037] at least one processor;

[0038] at least one memory for storing at least one program;

[0039] When the at least one program is executed by the at least one processor, the at least one processor implements the ship dynamics model identification method as described above.

[0040] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the ship dynamics model identification method as described above.

[0041] The above-mentioned technical solution of the present invention has at least one of the following advantages or beneficial effects: the present application calculates the cumulative probability density matrix of different types of ship data, thereby converting the distance similarity in the Euclidean space between the various types of ship data into the empirical similarity in the probability space, thereby emphasizing the similarity between the sparse values ​​of the ship data, and then combines the radial basis kernel algorithm to calculate the kernel function of the data pairs of multiple sample data, so that the high-dimensional fitting surface will not have too many abrupt values, making the identification of future ship dynamics models more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of a ship dynamics model identification method provided by an embodiment of the present invention;

[0043] Figure 2 is a schematic diagram of the comparison of similarity of sparse values ​​of normally distributed data provided by an embodiment of the present invention;

[0044] Figure 3 It is a flow chart of a ship dynamics model identification method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar components or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0046] In the description of the present invention, it should be understood that descriptions involving orientation, such as up, down, left, right, etc., the orientations or positional relationships indicated are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0047] In the description of the present invention, if there is a description of first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the sequence of the indicated technical features.

[0048] The embodiment of the present invention provides a ship dynamics model identification method, referring to Figure 1The ship dynamics model identification method of the embodiment of the present invention includes but is not limited to step S110, step S120, step S130 and step S140.

[0049] Step S110, obtaining a plurality of sample data in different time dimensions, wherein the sample data includes a plurality of ship data, and the plurality of ship data are divided into ship state quantity data and ship control quantity data;

[0050] Step S120, determining a cumulative probability density matrix of different types of ship data according to the plurality of sample data;

[0051] Step S130, determining a kernel function according to the cumulative probability density matrix of the plurality of ship data based on a radial basis kernel algorithm;

[0052] Step S140, training a least squares support vector machine according to the kernel function and the label data in the sample data to obtain a ship dynamics model, wherein the ship dynamics model is used to predict the ship state quantity data at the next moment.

[0053] In some embodiments, the ship is equipped with two global positioning systems (GPS) and an inertial navigation system (INS), and the ship position data and heading data are obtained through the two GPSs, and the inertial data can be obtained through the INS. The Kalman filter fusion algorithm is used to obtain various ship status data after the fusion of the two sets of GPS and INS data. The ship status data can be the longitudinal speed u, the transverse speed v and the bow angular velocity r. The stepper motor can control the steering and propeller rotation, and the ship control quantity data is obtained by obtaining the feedback data of the stepper motor. The ship control quantity data can be the rudder angle δ and the propeller speed n.

[0054] In some embodiments, ship data such as ship status data and ship control data during navigation are collected in real time to obtain sample data based on different moments. Each sample data includes multiple ship data. For example, the sample data X at the i-th moment is i =[u(i),v(i),r(i),δ(i),n(i)]. It can be understood that the sample data may also include more or less ship status data, ship control quantity data, or include other ship status data, ship control quantity data, and the embodiment of the present invention does not make specific limitations.

[0055] In some embodiments, the same type of ship data is extracted from multiple sample data and combined to obtain multiple different types of ship data sets, for example, a surge velocity data set, a sway velocity data set, a bow angular velocity data set, a rudder angle data set, and a propeller speed data set.

[0056] In some embodiments, in the initial stage of data analysis, the underlying structure of the data probability space is unclear, so the embodiments of the present invention use the empirical distribution of the observations to reflect the similarity between observation pairs. According to the analysis results, sparse values ​​have stronger similarity, that is, the data pairs at the center of the data space have stronger similarity than the data pairs at the edge at the same distance. For example, referring to Figure 2 , based on a set of normally distributed data, the explanation of sparse values ​​has stronger similarity. Sparse values ​​are data located on both sides with fewer occurrences. Figure 2 It can be seen that although the difference between the data pairs 4 and 3 and the data pairs 0 and 1 is 1, it is obvious that 4 and 3 are more similar than 0 and 1. Based on the above analysis, the data of various types of ships are preprocessed to determine the cumulative probability density of the ship data, and the cumulative probability density is further combined with the radial basis kernel function to train the least squares support vector machine, so as to obtain an accurate ship dynamics model. That is, the advantages of the Semblance kernel function and the RBF kernel function are combined to convert the distance similarity in the Euclidean space into the empirical similarity in the probability space, which can not only emphasize the similarity between sparse values, but also make the kernel function calculation more stable, that is, there will not be too many abrupt values ​​for high-dimensional fitting surfaces.

[0057] Specifically, assuming that sample data at 100 moments are collected, taking 100 surge velocities in the surge velocity data set as an example, the calculation process of the cumulative probability density of the surge velocity is:

[0058] Arrange the 100 surge velocities in the surge velocity data set from small to large, and the obtained surge velocities need to be recorded as data_1 to data_100. Then the cumulative probability density P_CDF_u of any surge velocity can be calculated by the following formula:

[0059]

[0060] Wherein, s represents the sequence number of the surge velocity.

[0061] If there are two identical surge velocities in the surge velocity data set, when the above formula is used for calculation, the serial number i of such data is calculated as follows:

[0062]

[0063] Based on the cumulative probability densities of all surge velocities, a surge velocity cumulative probability density matrix corresponding to the surge velocity data set can be obtained.

[0064] In some embodiments, the radial basis kernel function evaluates the similarity of data pairs using Euclidean distance, and commonly used radial basis kernel functions include Gaussian, multi-quadratic function, and thin plate spline function. Taking Gaussian as an example, the original Gaussian radial basis kernel function is as follows:

[0065] k(x,x')=exp(-a||x-x'|| 2 );

[0066] Based on the above Gaussian radial basis kernel function, after calculating the cumulative probability matrix corresponding to different types of ship data sets, the Gaussian kernel function of the ship data based on the probability space can be calculated as follows:

[0067] k(x,x')=exp(-a(1-|P_CDF_x-P_CDF_x'|) 2 );

[0068] Among them, P_CDF_x-P_CDF_x' represents the dissimilarity between data pairs in the probability density matrix corresponding to the ship data set, while 1-|P_CDF_x-P_CDF_x'| represents the similarity between data pairs, and a is the preset kernel width parameter. In this process, the data is indirectly normalized.

[0069] Furthermore, for the multidimensional data in the sample including the surge velocity, sway velocity, pitch angular velocity, rudder angle and propeller speed, the kernel function calculation formula of the sample is as follows:

[0070]

[0071] Among them, k(x,x') represents the kernel function of the data pair x and data x' in the sample, P_CDF_x n -P_CDF_x' n Represents the degree of dissimilarity between corresponding data pairs in the cumulative probability density matrix of the nth type of ship data, represents the sum of the similarities between corresponding data pairs in the cumulative probability density matrix of N types of ship data, N represents the number of ship data types, a is the preset kernel width parameter, for example, in this embodiment, N is 5.

[0072] In some embodiments, a least squares support vector machine algorithm may be used to identify a three-degree-of-freedom ship dynamics model. Specifically, the least squares support vector machine regression function is as follows:

[0073]

[0074] Among them, k(x,x') represents the kernel function, m represents the number of sample data in the sample data, α i and b are parameters to be trained, parameter α i and b are solved by the following formula:

[0075]

[0076] Among them, C is the preset penalty factor, y i is the label data in the i-th sample data. The label data can be the pitch speed, sway speed and yaw angular velocity at the next moment. If the label data is the pitch speed at the next moment, the prediction model for identifying the pitch speed u is obtained. If the label data is the sway speed v at the next moment, the prediction model for identifying the sway speed v is obtained. If the label data is the yaw angular velocity r at the next moment, the prediction model for identifying the yaw angular velocity r is obtained. In the training and testing process, the input is x i =[u(i),v(i),r(i),δ(i),n(i)]. After obtaining the three prediction models, the motion of a ship with three degrees of freedom can be predicted.

[0077] According to some specific embodiments of the present invention, step S120 includes but is not limited to the following steps:

[0078] The same type of ship data from the sample data at different times is extracted to obtain the ship data set;

[0079] Determine the sequence number of the ship data in the ship data set according to the size order of each ship data in the ship data set;

[0080] The cumulative probability density of the ship data is determined according to the serial number of each ship data in the ship data set, and the cumulative probability density matrix of the ship data set is obtained.

[0081] According to some specific embodiments of the present invention, step S140 includes but is not limited to the following steps:

[0082] Determine that the label data is the longitudinal velocity;

[0083] A least squares support vector machine is trained according to the kernel functions of multiple data pairs and the swell velocities in multiple sample data to obtain a swell velocity prediction model;

[0084] Determine that the label data is the sway velocity;

[0085] A sway velocity prediction model is obtained by training a least squares support vector machine according to the kernel functions of multiple data pairs and the sway velocities in multiple sample data;

[0086] Determine that the label data is the yaw angular velocity;

[0087] The least squares support vector machine is trained according to the kernel functions of multiple data pairs and the sway velocities in multiple sample data to obtain the yaw angular velocity prediction model;

[0088] A three-degree-of-freedom ship dynamics model is determined according to the surge velocity, the sway velocity and the bow angular velocity.

[0089] On the other hand, an embodiment of the present invention further provides a ship dynamics model identification system, comprising:

[0090] The first module is used to obtain a number of sample data in different time dimensions, wherein the sample data includes a plurality of ship data, and the plurality of ship data are divided into ship state quantity data and ship control quantity data;

[0091] The second module is used to determine the cumulative probability density matrix of different types of ship data based on multiple sample data;

[0092] The third module is used to determine the kernel function according to the cumulative probability density matrix of multiple ship data based on the radial basis kernel algorithm;

[0093] The fourth module is used to train a least squares support vector machine according to the kernel function and the label data in the sample data to obtain a ship dynamics model, wherein the ship dynamics model is used to predict the ship state data at the next moment.

[0094] It can be understood that the contents of the above-mentioned ship dynamics model identification method embodiment are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above-mentioned ship dynamics model identification method embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned ship dynamics model identification method embodiment.

[0095] Reference Figure 3 , Figure 3 Schematic diagram of a ship dynamics model identification device provided by an embodiment of the present invention. The ship dynamics model identification device of the embodiment of the present invention includes one or more control processors and a memory. Figure 3 A control processor and a memory are taken as an example.

[0096] The control processor and the memory can be connected via a bus or other means. Figure 3 The example of connecting through bus is taken in the following.

[0097] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the control processor, and these remote memories may be connected to the ship dynamics model identification device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0098] Those skilled in the art will understand that Figure 3The device structure shown in the figure does not constitute a limitation on the ship dynamics model identification device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0099] The non-transient software program and instructions required to implement the ship dynamics model identification method applied to the ship dynamics model identification device in the above embodiment are stored in the memory, and when executed by the control processor, the ship dynamics model identification method applied to the ship dynamics model identification device in the above embodiment is executed.

[0100] In addition, an embodiment of the present invention also provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are executed by one or more control processors, so that the one or more control processors can execute the ship dynamics model identification method in the above method embodiment.

[0101] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0102] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A ship dynamics model identification method, characterized in that: The following steps are involved: Acquire a plurality of sample data at different time dimensions, wherein the sample data includes a plurality of ship data, and the plurality of ship data are divided into ship state quantity data and ship control quantity data; Determine a cumulative probability density matrix of different types of ship data based on the plurality of sample data; Based on the radial basis kernel algorithm, a kernel function is determined according to the cumulative probability density matrix of the plurality of ship data; A least squares support vector machine is trained according to the kernel function and the label data in the sample data to obtain a ship dynamics model, wherein the ship dynamics model is used to predict the ship state quantity data at the next moment; Wherein, determining the cumulative probability density matrix of different types of ship data according to the plurality of sample data comprises the following steps: The same type of ship data from the sample data at different times is extracted to obtain the ship data set; Determine the sequence number of each ship data in the ship data set according to the size order of each ship data in the ship data set; Determine the cumulative probability density of the ship data according to the serial number of each ship data in the ship data set, and obtain the cumulative probability density matrix of the ship data set; The kernel function of the data pair in the kernel function matrix is ​​expressed as: Among them, k(x,x') represents the kernel function of the data pair x and data x', P_CDF_x n -P_CDF_x' n It represents the dissimilarity between corresponding data pairs in the cumulative probability density matrix of the nth type of ship data, N represents the number of ship data types, and a is the preset kernel width parameter; The step of training a least squares support vector machine according to the kernel function and the label data in the sample data to obtain a ship dynamics model comprises the following steps: Determine that the label data is the surge velocity; Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the surge velocities in the plurality of sample data to obtain a surge velocity prediction model; Determine that the label data is the sway velocity; A sway velocity prediction model is obtained by training a least square support vector according to the kernel function of the plurality of data pairs and the sway velocity in the plurality of sample data; Determine that the label data is the yaw angular velocity; Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the sway velocities in the plurality of sample data to obtain a yaw angular velocity prediction model; A three-degree-of-freedom ship dynamics model is determined according to the surge velocity prediction model, the sway velocity prediction model and the bow angular velocity prediction model.

2. The ship dynamics model identification method according to claim 1, characterized in that: The ship data includes surge speed, sway speed, bow angular velocity, rudder angle and propeller speed, wherein the surge speed, sway speed and bow angular velocity are ship state quantity data, and the rudder angle and propeller speed are ship control quantity data.

3. The ship dynamics model identification method according to claim 2, characterized in that: The least squares support vector machine is expressed as: Among them, k(x,x') represents the kernel function, m represents the number of sample data in the sample data, α i and b are parameters to be trained.

4. The ship dynamics model identification method according to claim 3, characterized in that: Parameter α i and b are solved by the following formula: Among them, C is the preset penalty factor, y i is the label data in the i-th sample data, and the label data is the ship state data at the next moment.

5. A ship dynamics model identification system, characterized in that: include: The first module is used to obtain a plurality of sample data in different time dimensions, wherein the sample data includes a plurality of ship data, and the plurality of ship data are divided into ship state quantity data and ship control quantity data; The second module is used to determine the cumulative probability density matrix of different types of ship data according to the plurality of sample data; A third module is used to determine a kernel function according to the cumulative probability density matrix of the plurality of ship data based on a radial basis kernel algorithm; The fourth module is used to train a least squares support vector machine according to the kernel function and the label data in the sample data to obtain a ship dynamics model, wherein the ship dynamics model is used to predict the state quantity of the ship at the next moment; Wherein, determining the cumulative probability density matrix of different types of ship data according to the plurality of sample data comprises the following steps: The same type of ship data from the sample data at different times is extracted to obtain the ship data set; Determine the sequence number of each ship data in the ship data set according to the size order of each ship data in the ship data set; Determine the cumulative probability density of the ship data according to the serial number of each ship data in the ship data set, and obtain the cumulative probability density matrix of the ship data set; The kernel function of the data pair in the kernel function matrix is ​​expressed as: Among them, k(x,x') represents the kernel function of the data pair x and data x', P_CDF_x n -P_CDF_x' n It represents the dissimilarity between corresponding data pairs in the cumulative probability density matrix of the nth type of ship data, N represents the number of ship data types, and a is the preset kernel width parameter; The step of training a least squares support vector machine according to the kernel function and the label data in the sample data to obtain a ship dynamics model comprises the following steps: Determine that the label data is the surge velocity; Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the surge velocities in the plurality of sample data to obtain a surge velocity prediction model; Determine that the label data is the sway velocity; A sway velocity prediction model is obtained by training a least square support vector according to the kernel function of the plurality of data pairs and the sway velocity in the plurality of sample data; Determine that the label data is the yaw angular velocity; Training a least squares support vector machine according to the kernel functions of the plurality of data pairs and the sway velocities in the plurality of sample data to obtain a yaw angular velocity prediction model; A three-degree-of-freedom ship dynamics model is determined according to the surge velocity prediction model, the sway velocity prediction model and the bow angular velocity prediction model.

6. A ship dynamics model identification device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the ship dynamics model identification method as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the ship dynamics model identification method as described in any one of claims 1 to 4 when executed by the processor.

Images